The NG2 proteoglycan is expressed by mural cells in both normal and pathological vasculature. NG2 is also expressed by many types of tumor cells. NG2 could therefore affect tumor growth and metastasis both by altering the intrinsic properties of tumor cells and by regulating tumor vascularization. Evidence exists for each of these possibilities. For example, NG2 expression by melanoma cells enhances their rate of proliferation, weakens their attachment to certain substrata, and promotes de-differentiation, resulting in faster tumor growth and enhanced metastasis. In the case of vascularization, NG2 expression by vascular mural cells appears to be required for the timely development of both macro- and microvascular structures, as evidenced by the finding that the NG2 knockout mouse exhibits delayed vascular morphcgenesis both pre- and postnatally. Since efficient vascularization is required for both tumor growth and metastasis, vascular NG2 could be a factor in regulating these aspects of tumor progression. This proposal will examine the role of both vascular and tumor cell NG2 in the vascularization and progression of tumors. Aim 1 will investigate the role of pericyte NG2 in the de novo development of breast and pancreatic tumors in MMTVIPyMT and RIP-tag transgenic mice, respectively. Introduction of these phenotypes onto the NG2 null background of NG2 knockout mice will allow assessment of the contribution of vascular NG2 to tumor progression. Aim 2 will evaluate the role of tumor cell NG2 in tumor progression. Comparison of NG2-positive and NG2-negative versions of the B 16 melanoma and the Lewis lung carcinoma will reveal NG2-dependent components of tumor growth and metastasis. Aim 3 will examine the role of NG2 in endothelial cell/mural cell communication. Co-culturing vascular endothelial cells along with NG2-positive or NG2-negative mural cells will allow a determination of the contribution of NG2 to cellular cross-talk that leads to the formation of vascular tubes.